DESCRIPTION: (Adapted from applicants abstract) Because approximately one million patients are tube-fed annually in the United States, nurses are frequently challenged to check feeding tube placement. Accurate placement of feeding tubes is key to ensuring adequate nourishment of patients and avoiding serious, if not catastrophic, pulmonary complications (such as feeding through a tube inadvertently positioned in the respiratory tract, or pulmonary aspiration in high-risk patients). At present, radiography is the only consistently reliable method for determining tube location. Unfortunately, repeated x-rays are expensive and pose a radiation risk; also, they are not readily accessible to patients in non-acute care settings. The proposed study is designed to determine the extent to which results from a combination of simple, inexpensive tests can be used at the bedside (in conjunction with a handheld calculator preprogrammed with logistic regression equations) to accurately predict tube location. These bedside tests involve: measuring the aspirate's pH; testing for the presence and concentration of trypsin, pepsin, and bilirubin; and examining the aspirate's appearance. (Abdominal x-rays will be taken at the time of collection of all gastrointestinal samples to serve as the "gold standard" against which the results will be compared.) The bedside tests have undergone initial testing concurrently with laboratory assays on split samples (at the time of abdominal x-rays to determine actual tube position); preliminary findings indicate high levels of sensitivity and specificity of the tests, both individually and collectively. In the proposed three-segment study, an additional 235 samples will be obtained by the investigators to allow refinement of the logistic regression equations needed to (1) differentiate between gastrointestinal and respiratory placement, and (2) gastric and intestinal placement. In the second segment of the study, 20 staff nurses will be instructed on the use of the bedside tests and calculator; after suitable rater agreement has been established, they will perform tests on 300 samples of gastric, intestinal, and respiratory fluids (randomly provided by the investigators) and predict tube location by use of the calculator. In the final segment of the study, the same 20 staff nurses will use the bedside tests and calculator to predict tube location at the time of 200 tube insertions in clinical settings. The extent to which these individuals can accurately predict feeding tube location by the proposed methods will be evaluated. The ultimate goal of the study is to provide clinicians with simple, inexpensive bedside tests (used in conjunction with a preprogrammed calculator) that can significantly increase the probability of accurately predicting feeding tube location, thereby decreasing reliance on costly, inconvenient, and potentially harmful x- rays. The preprogrammed calculator will serve as a decision-support tool in predicting feeding tube placement.